Solid-Phase Enzymatic Synthesis of Oligonucleotides - American

Aug 20, 1999 - antisense oligonucleotide concept.2 Originally, oligonucle- otide syntheses were carried out enzymatically,3 inspite of a number of dra...
2 downloads 0 Views 87KB Size
ORGANIC LETTERS

Solid-Phase Enzymatic Synthesis of Oligonucleotides†

1999 Vol. 1, No. 11 1729-1731

Carole Schmitz and Manfred T. Reetz* Max-Planck-Institut fu¨ r Kohlenforschung, Kaiser-Wilhelm-Platz 1, D-45470 Mu¨ lheim/Ruhr, Germany [email protected] Received August 20, 1999

ABSTRACT

The controlled and selective synthesis of oligonucleotides on the solid phase is possible under mild aqueous conditions using the enzyme T4 RNA ligase, the resins being tentagel or kieselguhr/polydimethylacrylamide.

Synthetic oligonucleotides play a key role in molecular biology and in chemistry.1 Prominent examples pertain to their use as probes for genes, as components in proteinDNA interactions, as tools in molecular cloning of DNA, and as building blocks in pharmaceuticals based on the antisense oligonucleotide concept.2 Originally, oligonucleotide syntheses were carried out enzymatically,3 inspite of a number of drawbacks such as laborious purification and the need to prepare protected nucleotides. Early enzymatic syntheses carried out on solid supports were abandoned due to an unacceptably low coupling yield.4 Following the advent †

Dedicated to Gu¨nther Wulff on the occasion of his 65th birthday. (1) See, for example: (a) Glick, B. R.; Pasternak, J. J. Molecular Biotechnology: Principles and Applications of Recombinant DNA, 2nd ed.; ASM: Washington, D.C., 1998. (b) Uhlmann, E.; Peyman, A. Chem. ReV. 1990, 90, 543-584. (c) Uhlmann, E.; Peyman, A.; Will, D. W. In Encyclopedia of Cancer; Bertino, J. R., Ed.; Academic: San Diego, 1996; Vol. 1, pp 64-81. (d) Fidanza, J. A.; Ozaki, H.; McLaughlin, L. W. In Protocols for Oligonucleotide Conjugates; Agrawal, S., Ed.; Methods in Molecular Biology (Totowa, New Jersey) 26; Humana: New Jersey, 1994; pp 121-143. (2) Reviews: (a) de Mesmaeker, A.; Ha¨ner, R.; Martin, P.; Moser, H. E. Acc. Chem. Res. 1995, 28, 366-374. (b) Thuong, N. T.; He´le`ne, C. Angew. Chem. 1993, 105, 697-723; Angew. Chem., Int. Ed. Engl. 1993, 32, 666. (c) Agrawal, S.; Iyer, R. P. Curr. Opin. Biotechnol. 1995, 6, 1219. (3) Examples: (a) Gilham, P. T.; Mackey, J. K. Nature (London) 1971, 233, 551-553. (b) England, T. E.; Uhlenbeck, O. C. Biochemistry 1978, 17, 2069-2076. (c) Hinton, D. M.; Brennan, C. A.; Gumport, R. I. Nucleic Acids Res. 1982, 10, 1877-1894. (d) Schott, H.; Schrade, H. Eur. J. Biochem. 1984, 143, 613-620. 10.1021/ol990240n CCC: $18.00 Published on Web 11/06/1999

© 1999 American Chemical Society

of the phosphoramide methodology, chemical synthesis became the standard protocol.1,2,5 Nevertheless, a few drawbacks remain, including the use of toxic reagents and solvents, the necessity of synthesis under an inert atmosphere, and purification problems. We now propose an oligonucleotide synthesis on the solid phase using the enzyme T4 RNA ligase,6 a process which offers a number of advantages by operating in a mild aqueous medium. The effectiveness of T4 RNA ligase in oligonucleotide synthesis is limited by the fact that it does not work equally well with a variety of acceptor and donor molecules.3b,7 Indeed, the formation of a 3′,5′-phosphodiester bond requires a trinucleoside biphosphate as a minimal acceptor and a mononucleoside 3′,5′-biphosphate as a donor. We therefore first prepared and purified by HPLC several 3′,5′-biphosphate (4) (a) Krynetskaya, N. F.; Tumanov, Y. V.; Potapov, V. K.; Shabarova, Z. A.; Prokof’ev, M. A. Dokl. Akad. Nauk SSSR 1981, 258, 1242-1245. (b) Mudrakovskaya, A. V.; Yamkovoy, V. I. Bioorg. Khim. 1991, 17, 819822. (5) For reviews, see: (a) Beaucage, S. L.; Iyer, R. P. Tetrahedron 1993, 49, 1925-1963. (b) Oligonucleotide Synthesis: A Practical Approach; Gait, M. J., Ed.; IRL: Oxford, 1984. (6) (a) Uhlenbeck, O. C.; Gumport, R. I. In The Enzymes, 3rd ed.; Boyer, P. D., Ed.; Academic: New York, 1982; Vol. 15 (Part B), pp 31-58. (b) Hinton, D. M.; Baez, J. A.; Gumport, R. I. Biochemistry 1978, 17, 50915097. (7) Romaniuk, E.; McLaughlin, L. W.; Neilson, T.; Romaniuk, P. J. Eur. J. Biochem. 1982, 125, 639-643.

donors 2 from the corresponding nucleosides 1 using the procedure of Barrio.8

In the next step, adenylated 3′,5′-deoxycytidine phosphate (4), although accessible by the method of Hoffmann and McLaughlin,9 was prepared more conveniently by reacting the (n-Oct)3NH+ salt of 2a with adenosine 5′-monophosphomorpholidate 3. Compound 4 was formed in 22% yield in addition to some 3′-adenylated deoxycytidine phosphate. Separation of this isomer was not necessary since it is not a substrate for the enzyme.

With these compounds and commercially available10 primers pdA5, 5′-pdA(pdA)3prC-3′, and 5′-pdA(pdA)9prC3′ in hand, we turned our attention to the solid phase, surmising that the choice of the solid support and of the linker could be crucial to the success of the enzymatic oligonucleotide synthesis. Indeed, only a limited number of successful cases of biocatalytic processes on solid supports are known based on the use of other enzymes and substrates.11 The commercial resins tentagel12 and kieselguhr/polydimethylacrylamide (PDMA)13 are available in forms having terminal NH2 groups, ideal for primer attachment. They appeared to (8) Barrio, J. R.; Barrio, M. del C.; Leonard, N. J.; England, T. E.; Uhlenbeck, O. C. Biochemistry 1978, 17, 2077-2081. (9) Hoffmann, P. U.; McLaughlin, L. W. Nucleic Acids Res. 1987, 15, 5289-5303. (10) EUROGENTEC, 4102-Seraing, Belgium. (11) Examples: (a) Elmore, D. T.; Guthrie, D. J. S.; Wallace, A. D.; Bates, S. R. E. J. Chem. Soc., Chem. Commun. 1992, 1033-1034. (b) Schuster, M.; Wang, P.; Paulson, J. C.; Wong, C.-H. J. Am. Chem. Soc. 1994, 116, 1135-1136. (c) Yamada, K.; Nishimura, I. Tetrahedron Lett. 1995, 36, 9493-9496. (d) Halcomb, R. L.; Huang, H.; Wong, C.-H. J. Am. Chem. Soc. 1994, 116, 11315-11322. (e) Ko¨pper, S. Carbohydr. Res. 1994, 265, 161-166. (f) Meldal, M.; Auzanneau, F.-I.; Hindsgaul, O.; Palcic, M. M. J. Chem. Soc., Chem. Commun. 1994, 1849-1850. (g) Waldmann, H.; Reidel: A. Angew. Chem. 1997, 109, 642-644; Angew. Chem., Int. Ed. Engl. 1997, 36, 647-649. (h) Sauerbrei, B.; Jungmann, V.; Waldmann, H. Angew. Chem. 1998, 110, 1187-1190; Angew. Chem., Int. Ed. 1998, 37, 1143-1146. (12) (a) Bayer, E. Angew. Chem. 1991, 103, 117-133; Angew. Chem., Int. Ed. Engl. 1991, 30, 113-129. (b) Rapp W. In Combinatorial Peptide and Nonpeptide Libraries; Jung, G., Ed.; VCH: Weinheim, 1996; pp 425464. (13) Dryland, A.; Sheppard, R. C. Tetrahedron 1988, 44, 859-876. 1730

fulfill the requirements with respect to insolubility, inertness to solvents/reagents, polarity, and sufficient surface area. Tentagel is based on a composite of cross-linked polystyrene with poly(ethylene glycol) (MW 3000-4000) and is terminally functionalized with NH2 groups. The beads swell in water, facilitating biological assays in aqueous systems. Because of their large loading capacity, they have been used for chemical oligonucleotide syntheses.14 The kieselguhrderived support consists of fabricated kieselguhr (SiO2) containing large pores in which cross-linked polydimethylacrylamide resin resides. Because of the macroporous structure of the pores, they are freely permeable for very large molecules and should be suitable for enzymatic synthesis. Our major efforts were therefore directed toward the use of this resin. Several 5′-phosphorylated primers (mainly commercially available) were coupled to the resins at the free amino group by performing the reaction in 0.1 M N-methylimidazole buffer (pH 6) in the presence of watersoluble 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), affording the resin-primer conjugate 5 having a loading of about 2 µmol/g as estimated from the UV analysis of the relevant solution following standard cleavage with NH2OH/ H2O (pH 4.5)14 and from analysis of the products by capillary electrophoresis (CE). Higher loading was not striven for, but there is no reason to assume that this is not possible.

The principle of our solid-phase enzymatic oligonucleotide synthesis is summarized in Scheme 1.

Scheme 1.

Principle of Solid-Phase Enzymatic Oligonucleotide Synthesis

The first three steps are repeated as often as desired, the final step being the cleavage from the resin. Specifically, the primer attached with its 5′-end to the solid support has a free 3′-OH group that binds the nucleoside 3′,5′-biphosphate in the presence of T4 RNA ligase. The terminal phosphate blocking group is removed enzymatically by alkaline phosphatase, and all excess reagents are simply washed off from the resin. The next nucleoside 3′,5′(14) Ryabova, T. S.; Shabarova, Z. A.; Prokof’ev, M. A. Dokl. Akad. Nauk SSSR 1965, 162, 1068-1070. Org. Lett., Vol. 1, No. 11, 1999

biphosphate can then bind to the end of the chain. The ligation reaction was carried out under two different reaction conditions (A and B). First, the process was carried out on a series of resin-supported primers using the ATP regeneration system (condition A) and different biphosphates. Fol-

Scheme 2.

NH2OH-Induced Selective Cleavage of Phosphoramide Bonds

solution and confirm the expectation that ribonucleotides are better acceptors than deoxyribonucleotides. NH2OH-mediated cleavage sets free the oligonucleotide as detected by CE/ MALDI-TOF-MS. At long reaction times, none of the primer was detected, indicating a high coupling efficiency of the ligation. Furthermore, RNase A selectively cleaves the synthesized oligonucleotide at the ribose sugar,15 in our case the primer remaining on the resin (Scheme 3; Table 3). In

Scheme 3.

lowing NH2OH-induced cleavage from the resins (Scheme 2), the desired products were obtained as shown by CE (Table 1).

Table 1. Solid-State Enzymatic Oligonucleotide Synthesis under Conditions Aa solid phase

primer

pdNp

final product

kieselguhr/PDMA kieselguhr/PDMA kieselguhr/PDMA kieselguhr/PDMA kieselguhr/PDMA tentagel

pdA4prC pdA4prC pdA4prC pdA4prC pdA10prC pdA4prC

pdCp pdAp pdGp pCp pdCp pdCp

pdA4prCpdCp pdA4prCpdAp pdA4prCpdGp pdA4prCpCp pdA10prCpdCp pdA4prCpdCp

aTris HCl 50 mM pH 8, MnCl 10 mM, DTT 20 mM, ATP 0.4 mM, 2 phosphocreatine 5 mM, spermine 8mM, myokinase 170 U/mL, creatine kinase 175 U/mL, BSA 10 µg/mL, pdNp 8 mM, S-pdN(pdN)n 0.5 mM, T4 RNA ligase 3300 U/mL, 17 °C/48 h. Cleavage: NH2OH/H2O (1:1), pH 4.5; 1 h/37 °C.

The ligation reaction using the pre-adenylated 2′-deoxycytidine phosphate was also carried out, specifically under conditions B which were also applied in the coupling of a second nucleoside 3′,5′-biphosphate (Table 2).

Table 2. Solid-State Enzymatic Oligonucleotide Synthesis under Conditions Ba solid phase

primer

kieselguhr/PDMA kieselguhr/PDMA kieselguhr/PDMA kieselguhr/PDMA

pdA5 pdA4prC pdA10prC pdA10prC

reaction AppdNp time (h) AppdCp AppdCp AppdCp AppdCp

144 48 48 96

final product pdA5pdCp pdA4prCpdCp pdA10prCpdCp pdA10prCpdCpdCp

a Tris HCl 50 mM pH 8, MnCl 10 mM, DTT 20 mM, BSA 10 µg/mL, 2 AppdNp 5 mM, S-pdN(pdN)n 0.5 mM, T4 RNA ligase 3300 U/mL, 17 °C. Cleavage: NH2OH/H2O (1:1), pH 4.5; 1 h/37 °C.

All reactions proceeded to completion and were monitored by CE, including the cleavage processes. The results show that the composition of the primer has an effect on the reaction rate. For the primer with a ribocytidine at the end, the reaction is complete after 48 h, whereas the use of a pure deoxyribonucleotide primer requires 144 h for completion of reaction. These values are comparable to those in Org. Lett., Vol. 1, No. 11, 1999

RNase A Catalyzed Cleavage

this case the products were analyzed by CE and identified by comparison with commercially available nucleoside 3′phosphates.

Table 3. Products of RNase A Catalyzed Cleavage from Kieselguhr-PDMA Bound Nucleotides

Finally, the extended primer attached to the resin can be treated with alkaline phosphatase16 to give the dephosphorylated product, the free 3′ end group allowing for attachment of a second nucleoside 3′,5′-biphosphate if so desired. This can be conveniently carried out if the enzyme is deactivated by heat treatment (95 °C/5 min) prior to the coupling process. In summary, our nonoptimized preliminary results clearly show that T4 RNA ligase, although somewhat slow, principally catalyzes under mild conditions the attachment of various 3′,5′-diphosphates and the pre-adenylated 3′-nucleoside phosphate to primers immobilized on solid supports. The relatively low rate is not due to possible immobilization effects. Rather, T4 RNA ligase is known to be a fairly slow enzyme in solution.3b,7 In view of the possibility of automation and molecular biological optimization of enzymes using directed evolution,17 the enzymatic synthesis of oligonucleotides on the solid phase may turn out to be a viable alternative to current conventional chemical methods. Supporting Information Available: Typical experimental procedures. This material is available free of charge via the Internet at http://pubs.acs.org. OL990240N (15) Richards, F. M.; Wyckoff, H. W. In The Enzymes, 3rd ed.; Boyer, P. D., Ed.; Academic: New York, 1971; Vol. 4, pp 647-806. (16) Kunitz, M. J. Biol. Chem. 1946, 164, 563-568. (17) (a) Arnold, F. H. Acc. Chem. Res. 1998, 31, 125-131. (b) Stemmer, W. P. C. Nature (London) 1994, 370, 389-391. (c) Reetz, M. T.; Jaeger, K.-E. Top. Curr. Chem. 1999, 200, 31-57. 1731